BetaEntity Annotation Prototype
← Back to diseases

Annotated abstract

M39 A piece of the diagnostic puzzle: could histological results be used to trigger a specialist TB review?

thoraxjnl · 2025-11-02 · canonical JSON source

2 visible annotations · policy: published · automated confidence ≥ 75.00%

Document resource

Background In 2024 5,480 people were treated for TB disease in England. 1 Early diagnosis and treatment reduces disease-associated morbidity and mortality. A typical histological pattern of necrotising granulomatous inflammation is key diagnostically, due to long delays in positive culture results. We reviewed all biopsies coded with this finding to determine whether active appraisal of these patient cases could reduce diagnostic delay of TB.Methods We undertook a retrospective cohort analysis of all biopsy reports of ‘necrotising granulomatous inflammation’ from April 2023–2024 at two TB services in an urban area. Data were collected from the electronic patient health record and pathology database and cross-referenced with the national TB database to verify patients diagnosed with active TB.Results 64 cases were found; 18 had reports where the search term was mentioned as a relevant negative finding in the histology report. Of the remaining 46 cases, 4 had a known cause for the necrotising granulomatous inflammation and they were excluded, resulting in a total of the 42 cases that were studied in further detail.Patient characteristics are referenced in table 1.In the TB patient cohort, median time from biopsy result to treatment initiation was 2 days (IQR 0 – 47 days). 9 patients received treatment within 7 days of biopsy result; of the remaining 5 patients, median time from result to treatment initiation was 91 days (IQR 27–109 days).Abstract M39 Table 1 Whole cohort (n=42) TB diagnosis (n=14) Not TB (n=28) Age [years] median (IQR) 57.4 (44.7 – 70.0) 44.6 (35.0–59.0) 57.6 (52.6–74.6) SexMaleFemale 18 (43%)24 (57%) 6 (43%)8 (57%) 12 (43%)16 (57%) Biopsy sitePulmonary (%)Lymph node (%)Soft tissue (%) 10 (23%)12 (29%)20 (48%) 4 (28%)5 (36%)5 (36%) 6 (21%)5 (18%)17 (61%) Acid fast bacilli Positive Negative Non-diagnostic 6 (14%)25 (60%)11 (26%)6 (43%)8 (57%)016 (57%)12 (43%) Necrosis present on sample Positive Negative 25 (60%)17 (40%) 14 (100%)11 (39%)17 (61%)Discussion We suggest that active review of histology samples reported as ‘necrotising granuloma’ may reduce diagnostic delay. A significant proportion of patient biopsies with this finding were subsequently diagnosed with TB, and the overall numbers do not suggest that a streamlined pathway prompting review of the patient notes within the TB service would be a strain on resources. Prolonged time to diagnosis can have consequences for the individual, the health service and the public; a fast-track system directing pathology reports with key phrases to the TB service could translate into measurable benefit, and help avoid detrimental delay.Reference National quarterly report of tuberculosis in England: quarter 4, 2024, provisional data - GOV.UK